Inflammatory bowel disease (IBD) includes a group of diseases, the most common of which are Crohn's disease and ulcerative colitis. Although these diseases are not frequently encountered, most pediatricians will treat patients with IBD at some point in their career, and inevitably will be faced with an extremely complex somatic and psychiatric clinical picture (1). Although regarded as separate clinical entities, Crohn's disease and ulcerative colitis share many clinical characteristics. The course of these diseases is usually lifelong with exacerbations and remissions and is therefore unpredictable.
Many studies have shown that chronic diseases in general predispose children to psychiatric disorders. In a meta-analytic review of different chronic diseases, IBD seems to have the most profound effect on mental health of all medical diseases reviewed (2). In the absolute majority of studies, more than 50% of children with IBD have had psychiatric disorders according to the classifications provided by the Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition (DSM-III-R). Because the IBD group differs significantly from children with other diseases, it seems clear that this is not a general effect from having a chronic disease; instead, it is more specifically related to the particular clinical features of IBD. In most studies, there are no clear differences between ulcerative colitis and Crohn's disease with respect to psychiatric disorders.
A majority of the controlled studies of children and adolescents with IBD have reported high rates of psychiatric symptoms. In one study, 56% of the children with IBD compared to 18% in the control group had a psychiatric disorder, almost exclusively emotional disorders (3). Other studies have found depressive disorders and lower self-esteem to be more common in children with IBD than in healthy control subjects (4,5). Obsessive-compulsive disorders have also been found to be common (6). Behavioral problems have been found to be significantly more common in children and adolescents with IBD as compared to their siblings (7).
MATERIALS AND METHODS
The physical and mental health of a consecutive series of children and adolescents in Örebro county, Sweden, was examined with a multimethod design (8,9). The study also aimed to investigate the correlations between psychiatric problems and severity levels of the disease. Another goal of the project was to gather information concerning salutogenetic factors that could explain why some children eventually may come out well psychologically despite a complex medical situation.
All children and adolescents with IBD in the age range of 7 to 18 years were invited to participate in the study. Twenty (83%) of the 24 families eligible for the study agreed to participate. Several different comparison groups were examined to broaden the possibilities of comparison. All comparison groups were individually matched to the IBD group for age, sex, and choice of subjects in school. Two clinical contrast groups, one with diabetes (D) and the other with chronic tension headache (H), were evaluated. A group of physically healthy children and adolescents (HC) matched with the IBD group was also examined.
The study thus comprised 80 subjects, 20 in each group, aged 9 to 18. The mean age in the IBD group and in the comparison group was 16.5 years (9-16). There were 9 boys and 11 girls. There were only small differences regarding ordinal position among siblings, school situation, parent's age, occupation, and family income.
In the IBD group, there were 11 subjects with ulcerative colitis and 9 with Crohn's disease. The mean age at onset was 12.2 years (range, 7-16) and the duration of the disease was 4.3 years (range, 1.0-8.2).
All children were interviewed in an in-depth manner using the Child Assessment Schedule (10). This interview, lasting 1 to 3 hours, yielded a diagnosis according to the DSM-III-R classification of mental disorders. The children with IBD were also interviewed specifically about coping strategies used to handle their disease, the effect of the disease on everyday life, and their thoughts about how they could possibly affect the course of the disease. The interview also related to potential salutogenetic factors. This material was collected using a specific semistructured interview schedule.
The children completed several questionnaires: the Frisk well-being scale (Frisk M. The Frisk well-being scale for children. Unpublished manuscript. Uppsala, Sweden: Department of Child and Adolescent Psychiatry, Uppsala University, 1986 [in Swedish]), the Children's Depression Inventory (Kovacs M. The Children's Depression Inventory: A self-rated depression scale for school-aged youngsters. Unpublished manual. Pittsburgh, PA: University of Pittsburgh, 1982), the Revised Children's Manifest Anxiety Scale (11), the Jag Tycker Jag Är ("I Think I Am") (12), the Locus of Control Scale (13), the Rotter Sentence Completion Test (14), and Raven's Matrices (15).
The children's mothers were interviewed about the effect of the disease using a semistructured interview schedule. They also completed the Child Behavior Checklist (16) and the Frisk well-being scale. Both parents completed the Symptom Checklist-90 (17), the Interview Schedule for Social Interaction (18), and the FANS questionnaire concerning family interaction (Cederblad M. FANS. Unpublished interview schedule. Lund: Department of Child and Adolescent Psychiatry, University of Lund, 1985 [in Swedish]).
The studies were approved by the Medical Research Ethics Committee of Örebro and Uppsala. The children and their families were informed that the aim of the study was to investigate the well-being of chronically ill children and their families. Informed consent was given by the children and their parents. They also agreed to provide access to all relevant medical files.
Some of the results have been previously published (8,9,19,20). In this article, a summary of these studies is presented with new results from the disease-related interview with the IBD group.
The Child Assessment Schedule interview revealed a prevalence of psychiatric disorder of 60% in the IBD group. Of these, half were considered to be severe or moderate disorders that influenced the patient's daily life. In the H group, 30% showed psychiatric disorders; in the D group, 20% showed psychiatric disorders; and in the HC group, 15% showed psychiatric disorders (p = 0.01). The disorders were not randomly spread but clustered clearly around depressive and anxiety disorders (Table 1).
The total behavior problem scores recorded on the Child Behavior Checklist, which provides a general picture of the children's psychological adjustment (according to the mothers), were higher in the IBD and H groups than in the D and HC groups, with a significant post hoc difference between the IBD and HC groups (p < 0.01) (Table 2). This was true of both the externalizing and the internalizing subscales, which tended to co-vary to a high degree.
The results of the Child Assessment Schedule and the Child Behavior Checklist were correlated to certain somatic variables in the IBD group. No significant correlations were found with duration of disease, total disease score, or the somatic status at the time of examination. This means that nonlinearity was found between the physical severity and psychiatric symptoms.
The self-esteem reported by the children on the Jag Tycker Jag Är questionnaire showed significant group differences (p < 0.04), with the IBD and H groups scoring lower than the HC group, and the D group being intermediate.
The Frisk well-being scale was completed by both the children and their mothers. The children showed significant group differences (p < 0.04), with the H group scoring generally lower than the HC group, and the IBD and D groups being intermediate. The mothers, on the other hand, provided a different picture. According to answers given by the mothers, the IBD group had significantly lower scores (p < 0.04) than the HC group, with the H and D groups being intermediate.
The results of the Children's Depression Inventory scale showed significant group differences (p < 0.02), with the IBD and H groups scoring higher than the HC group. On the Revised Children's Manifest Anxiety Scale, the results were more complex. For the overall anxiety level, the H group scored significantly higher than the other three comparison groups. However, "physiological anxiety" was the only subscale that showed significant group differences. The Lie scale included in the Revised Children's Manifest Anxiety Scale yielded significant group differences (p < 0.001), with the IBD group scoring much higher than all the comparison groups. The children in the IBD group scored significantly higher (p < 0.01) on the Locus of Control Scale than the D and HC groups. This means that they showed a more external locus of control than did the other children.
The correlation between the Locus of Control Scale and psychiatric disorders was very low (rS, = 0.01) but the scattergram revealed a curvilinear relationship (Fig. 1) between these two variables. Both high and low levels on the Locus of Control Scale seemed to correlate with severe psychiatric disorders.
The total scores on the Symptom Checklist-90 scale for current psychopathology in the mothers were significantly different between the IBD and HC groups (p < 0.001), with the mothers in the IBD group having much higher levels of psychopathologic symptoms. The scores of the fathers in the two groups were similar, and no significant group differences were detected.
The global social support scores (Interview Schedule for Social Interaction) were significantly different for both the mothers and the fathers (p < 0.05), with the parents in the IBD group scoring lower than the HC group. The subscales for social integration were similar in both groups, both on the quantitative measure (Availability of Social Interaction) and the qualitative measure (Adequacy of Social Interaction). On the parameters for attachment, however, several differences were found. The availability of attachment differed significantly, with the mothers (p < 0.01) and fathers (p < 0.01) of children with IBD scoring lower than the HC group. For the qualitative aspect, that is, the adequacy of attachment, the mothers of children with IBD scored significantly lower than the HC group (p < 0.05), whereas the fathers' results revealed only a tendency in the same direction.
The scores on the FANS questionnaire for the mothers showed significant differences between the groups (p < 0.01), those in the IBD group having scores that indicated higher degrees of family dysfunction than those in the D and HC groups. The fathers showed tendencies in the same direction as the mothers, but the differences were too small to be statistically significant. The variations were high in all three groups.
In the IBD group, the differences between the parents were small, and the scores of the fathers indicated a slightly lower level of family dysfunction than did those of the mothers. In both the D and the HC groups the differences were larger and in the opposite direction.
The disease-specific interviews in the IBD group were somewhat difficult to carry out because of the children's tendency to give only short and concrete answers that were found to be "emotionally thin." Some of them were reluctant to discuss their disease and denied that it had any significance in their daily life. The children in this subgroup were the ones who exhibited a combination of psychiatric disorder, high levels on the "Lie scale," and high degrees of external locus of control. For most of these children it was found that in their medical files, where psychological information otherwise is scarce, the children were described as "dissimulating?"
Another subgroup that had a fairly aggressive somatic illness but seemed to come out well in psychological and social terms was identified. This children in this group did not differ from the others with respect to sociodemographic background factors. They did, however, exhibit certain common characteristics.
In this subgroup, the knowledge of the disease and also the eagerness to learn more about it were obvious. These children also thought that they could, in one way or another, affect the course of their disease to a certain extent. They could all name factors in life that affected their disease in a negative way. These factors were often of a common, everyday nature but did have a stressful effect on the child. This kind of experience often pertained to exacerbations of symptoms like diarrhea or increased intestinal motility or pains, both in active periods of the disease and in periods of remission that could still give rise to intestinal symptoms.
In this subgroup, an open, permissive climate within the family with a well-functioning family social network was more common. The individuals in this subgroup also had someone to talk to concerning their reactions toward the disease. There was room to exhibit anxiousness, depression, anger, or despair. The person who was the listener was sometimes a family member, often the mother, but could in some cases be someone outside the family, for example, a person considered a "significant other."
The results of this study can be generalized only with caution, because the samples are limited. It should also be noted that the results are relevant only to school-aged children and adolescents. The high participation rate and the use of validated instruments nevertheless make possible an interpretation of the findings. In any study using a large number of measurements, statistically significant results might actually have occurred by chance. However, the differences detected formed a regular pattern across tests that is unlikely to be a chance finding alone.
The multimethod, multicomparison approach that was used to study the mental health of children with IBD showed that, on most of the variables, children with IBD had the highest levels of psychiatric disturbance among the groups studied. Thus, it is clear that children and adolescents with IBD constitute a population at high risk for psychiatric disorders.
The observed difference between IBD and the other clinical groups indicates that the higher risk of psychiatric disorder in the IBD group is probably not a general effect of having a chronic medical disease, but is more specifically related to the clinical features of IBD.
A possible explanation of the observed differences between the disease groups may be the character and nature of the symptoms. Diabetic children may experience many difficulties in their daily lives, but after their initial difficulties in accepting the role of a sick person, they realize that the disease is socially acceptable and gives no cause for shame. The symptoms of IBD are more often socially embarrassing and humiliating and are very hard to discuss. Most of the children with IBD who were interviewed said that they were not regarded as ill by their peers, and even in the family circle, they seldom spoke of the illness. This way of handling the disease may obviously produce symptoms, mainly depression and anxiety.
The coping responses used to handle a chronic disease may vary considerably. The concept of locus of control has proved to be a good predictor of management skills, achievement, and well-being in chronically ill children. The hypothesis that children with IBD might show a more external locus of control than the comparison groups was confirmed. The reason for this is not fully understood. Because IBD is highly unpredictable and the factors that may cause or aggravate it are unclear, an external locus of control would be predicted in these children.
The curvilinear relationship between locus of control and psychiatric disorder should be interpreted with caution because the sample is limited. It is tempting to speculate that the relation between locus of control and mental health is two-edged. The intuitive interpretation would be that an external locus of control, with feelings of helplessness, dependence, and tension, corresponds to poor mental health, whereas an active, responsible, and independent person with an internal locus of control enjoys better general mental health. However, there may be a "break point" at which this effect becomes negative. A high level of internality in the locus of control dimension may lead the subjects to feel guilt and blame the disease on themselves. An extremely high personal responsibility for a chronic disease may therefore be too heavy a burden for a child or adolescent during a period of life when questions of independence and self-reliance are crucial.
The families were also affected, often to a considerable extent. This was true especially of the mothers. However, there was no significant correlation between the mothers' distress and the psychiatric disorders of the children. This rather remarkable finding may be interpreted as a consequence of the tendency within these families to isolate feelings. The mother therefore has to bear her worries and concerns all alone with little possibility of sharing them with her husband. The mothers' constant worries about the development of the disease were not obvious to their children. This may be viewed as a kind of vicarious suffering.
A good social support system in the family seemed to fill a protective function for the children. Although the mental health of the parents per se did not seem to affect the children very much, it was obvious that any injury to the deeper relations within the family might have a profound influence on the children's mental health.
Treatment of IBD in children should always include both medical and psychological approaches. A careful explanation from the pediatrician about the joint medicopsychological approach is important in helping to reduce stigma and in preventing the family from feeling "blamed" for the illness. It is important to convey to the family that the psychological approach aims to achieve two goals. The first is to provide support for the child and the parents in coping with the serious disease. The second is to map, understand, and alleviate stresses that may hinder recovery or cause exacerbations of the disease.
It is of utmost importance to create an atmosphere that permits both the child and the parents to express their concerns about the physical and the psychological aspects of the disease. These studies have clearly shown that children with IBD experience difficulties in expressing emotions. It is therefore unlikely that they will share their worries and concerns with the medical staff at the beginning of treatment. As time goes by, a close and empathetic relationship between the medical team and the patient may enhance the likelihood of emotional expression.
It is a common experience that children with IBD show a tendency to deny their problems, in terms of both the symptoms of the disease and the coexisting psychological problems. They may exhibit a dissimulating style in communicating with their doctor. A high level of sensitivity for the message from the patient is therefore needed. Careful observation of nonverbal communications can assist in the recognition of difficulties previously not expressed.
It is important to see the child alone to make psychological communication easier. It is also important to educate the child about the disease, because a well-informed child may be more likely to develop active coping strategies. The problems that may be elicited from such an approach can be handled in several ways. The coping strategies that the child uses should be developed, and if they are found to be inappropriate, they should be discussed with the child.
Depressive disorders are commonly encountered in children and adolescents with IBD. These problems may be detectable at the beginning of the disease, but more often develop after some period of time, perhaps when the child realizes that he or she will have to live with a chronic disease. The depressive disorder should be dealt with psychotherapeutically, either with the child or with the whole family, depending on the kinds of problems seen.
The results of the study shed light on the complex interplay between physical, psychological, and social factors in the course of a chronic medical disease in children and adolescents. As one author (21) who has ulcerative colitis stated, "The fact that having chronic ulcerative colitis is as much a state of mind as a state of physical being, should be remembered by the doctor."
1. Engstrom I, Lindquist BL. Gastrointestinal disorders. In: Ammerman R, Campo J, Eds. Handbook of Pediatric Psychology and Psychiatry.
New York: Allyn & Bacon.
2. Lavigne J, Faier-Routman J. Psychological adjustment to pediatric physical disorders: a meta-analytic review. J Pediatr Psychol
3. Steinhausen HC, Kies H. Comparative studies of ulcerative colitis
and Crohn's disease
. J Child Psychol Psychiatry
4. Burke P, Meyer V, Kocoshis S, et al. Depression and anxiety in pediatric inflammatory bowel disease
and cystic fibrosis. J Am Acad Child Adolesc Psychiatry
5. Raymer D, Weininger O, Hamilton J. Psychological problems in children
with functional abdominal pain. Lancet
6. Burke P, Meyer V, Kocoshis S, Orenstein D, Chandra R, Sauer J. Obsessive-compulsive symptoms in childhood inflammatory bowel disease
and cystic fibrosis. J Am Acad Child Adolesc Psychiatry
7. Wood B, Watkins JB, Boyle JT, Nogueira J, Zimand E, Carroll L. Psychological functioning in children
with CD and ulcerative colitis
: Implications for models of psychophysiological interaction. J Am Acad Child Adolesc Psychiatry
8. Engstrom I, Lindquist BL. Inflammatory bowel disease
: A somatic and psychiatric investigation. Acta Paed Scand
9. Engstrom I. Mental health and psychological functioning in children
with inflammatory bowel disease
: A comparison with children
having other chronic illnesses. J Child Psychol Psychiatry
10. Hodges K, McKnew D, Cytryn L, Stern L, Kline J. The Child Assessment Schedule (CAS) diagnostic interview: A report on reliability and validity. J Am Acad Child Psychiatry
11. Castaneda A, McCandless BR, Palermo DS. The Children
's form of the Manifest Anxiety Scale. Child Dev
12. Ouvinen-Birgerstam P. Jag tycker jag är
. Manual. Stockholm: Psykologiförlaget, 1985 (in Swedish).
13. Nowicki S, Strickland BR. A locus of control scale for children
. J Consult Clin Psychol
14. Rotter B. The Rotter Incomplete Sentences Book.
Manual. New York: The Psychological Corporation, 1950.
15. Raven JC, Court JH, Raven J. Manual for Raven's Progressive Matrices and Vocabulatory Scales.
London: HK Lewis & Co. Ltd., 1983.
16. Achenbach TM. Manual of the Child Behavior Checklist (CBCL).
Burlington, VT: University Associates of Psychiatry, 1988.
17. Derogatis LR, Lipman RS, Covi L. SCL-90: An outpatient psychiatric rating scale. Psychopharmacol Bull
18. Henderson S, Byrne DG, Duncan-Jones P. Neurosis and the Social Environment.
Sydney: Academic Press, 1982.
19. Engstrom I. Parental distress and social interaction in families with children
with inflammatory bowel disease
. J Am Acad Child Adolesc Psychiatry
20. Engstrom I. The influence of family interaction and locus of control on coping processes in children
with inflammatory bowel disease
. J Am Acad Child Adolesc Psychiatry
21. Kelly MP. The subjective experience of chronic disease: Some implications for the management of ulcerative colitis
. J Chronic Dis
Manchester Business School, 17 September 1997
On behalf of the North American and European Collaborative Research Groups on Pediatric IBD